From dos at scarff.id.au Sun Jan 3 21:11:24 2021 From: dos at scarff.id.au (Dean Scarff) Date: Mon, 04 Jan 2021 13:11:24 +1100 Subject: [Cake] [Bloat] New board that looks interesting In-Reply-To: References: <1586011622.632930657@apps.rackspace.com> Message-ID: Any stats on how much power it pulled during your tests and when idle? On Fri, 18 Dec 2020 15:48:46 -0800, Aaron Wood wrote: > I have, finally.  It's been running for a week or so, now. > > OpenWRT was an _adventure_.  The board is UEFI, not standard bios..  > And while it will merrily boot OpenWRT's non-uefi images off of USB, > it won't boot the non-UEFI setup from the internal storage (I'm using > the eMMC).  So _that_ was fun (and I made some dumb mistakes that > were especially fun to correct. > > But it's running OpenWRT 19.07 (and a UEFI bootloader before grub > that's from ToT OpenWRT). > > Anyway, I have cake running, 950Mbps ingress and 35Mbps egress (modem > is provisioned at 1.3G ingress, and a bit over 35Mbps egress). >  fq_codel was defaulted, in multi-queue mode.  While I'm using cake > on ingress, my local link hasn't been hitting the limiter very often: > >                 Tin 0 >   thresh        950Mbit >   target          1.5ms >   interval       30.0ms >   pk_delay         22us >   av_delay          9us >   sp_delay          2us >   backlog            0b >   pkts        243608193 >   bytes    250748364896 >   way_inds     13167720 >   way_miss      1245030 >   way_cols            0 >   drops            1075 >   marks             101 >   ack_drop            0 >   sp_flows            0 >   bk_flows            1 >   un_flows            0 >   max_len         69876 >   quantum          1514 > > Given that most of the hosts that I interact with are only about > 10-15ms away, I'm probably going to change the interval target to > better match that. > > Interestingly, while it has a pair of multiqueue NICs (i211s), the > igbe driver isn't configuring them for RSS.  Both output queues are > being used, but not the ingress queues: > > wan interface: > >      tx_queue_0_packets: 56635989 >      tx_queue_1_packets: 39777210 >      rx_queue_0_packets: 243646072 >      rx_queue_1_packets: 0 > > lan interface: > >      tx_queue_0_packets: 85047897 >      tx_queue_1_packets: 162004500 >      rx_queue_0_packets: 111174855 >      rx_queue_1_packets: 0 > > Since I have housemates that don't appreciate me messing with the > network during their meetings, I haven't gotten around to poking more > deeply at that (or at experimenting with running cake on two ingress > queues). > > That being said, I bench-tested this before I put it into operation > and was able to see 940Mbps of iperf goodput through cake and NAT...  > Took all of a core, though (and that core was still cold and > therefore > potentially able to boost to 2.5GHz).  I haven't determined how long > it will take to thermally throttle, and if bandwidth suffers as a > result. > > Pretty happy with it so far, though.